CCSS: Autonomous Drone and Ground Robot Cooperative Tasking in Complex Indoor Environments

CCSS：复杂室内环境中的自主无人机和地面机器人协作任务

• 批准号: 1923163
• 负责人: Shiwen Mao
• 金额: $ 39.89万
• 依托单位: Auburn University
• 依托单位国家: 美国
• 项目类别: Standard Grant
• 财政年份: 2019
• 资助国家: 美国
• 起止时间: 2019-08-01 至 2023-07-31
• 项目状态: 已结题
• 来源: https://www.nsf.gov/awardsearch/showAward?AWD_ID=1923163&HistoricalAwards=false
• 关键词: CCSS; Autonomous; Drone; Ground; Robot

The requirements for automated inventory and precise location of items have become vital to modern supply chain management. The objective of this project is to create an innovative ground robot-drone network system, which consists of a group of autonomous ground robots and drones, to provide inventory counts and precise locations of passive radio-frequency identification (pRFID) tagged items in highly complex environments, such as warehouses, retail stores, hazmat storage facilities, or factories. This project will significantly improve the state-of-the-art of supply chain management and Internet of Things (IoT) systems, and provide a significant step forward to fully harvest the potential of the proposed robotic-drone platform. The project's education plan includes developing and enhancing various undergraduate and graduate-level courses. Graduate and undergraduate students will be exposed to the state-of-the-art techniques, and gain hands-on experience in the cutting-edge technology that is at the very frontier of modern communications, circuits, and sensing systems (CCSS). Outcomes from this project will be disseminated through technical publications, conference presentations, a project website, and at the bi-annual Wireless Engineering Research and Education Center (WEREC) and RFID Lab meetings. The team is fully committed to promoting participation from under-represented groups in research, and will continue such efforts via outreach, e.g., through the NSF REU and RET programs and collaboration with HBCUs.The pRFID technology has been widely deployed in the past decade for serialized item level identification and data sharing. However, most pRFID technology implementations utilize fixed reader points, or human operated handheld scanners, and cannot provide precise item location. The demand of logistics visibility requires automated inventory and the precise location information of items. In the proposed research, the autonomous ground robot-drone network system combined with a precise RFID localization method will bridge the above gap. By deploying cooperative ground robots and drones, mounted with commercial off-the-shelf pRFID equipment, to provide automated inventory and precise location of pRFID tagged items. The framework cooperates heterogeneous individual items into a coherent system for more complex task that is not possible for any individual item. The proposed architecture will also provide an innovative communication, control, and computing framework for general IoT systems. The framework will be disclosed as open-source tools to boost relevant research in the CCSS community. The following thrusts will be accomplished in this project. (i) Ground robot and drone network architecture: the architecture will be developed to provide communication, computing, and control for the ground robot and drone to cooperatively operate for generic tasks. It also provides the fundamental methods for the ground robot and drone to pair with each other to form a symbiotic system. (ii) Ground robot and drone indoor navigation: a ground robot enhanced mechanism will be introduced to enable drone(s) to precisely localize itself in the complex indoor environment. When the localization goals are achieved, a method will be investigate to enable the drone(s) and ground robot to safely and efficiently navigate in an object-rich and confined space environment. (iii) Accurate inventory counts and precise localization of pRFIDs: the ground robot-drone network will be prototyped to operate pRFID inventory and provide precise location of pRFID tagged items. (iv) This project also includes a thorough integration and assessment plan, to test the proposed ground robot-drone integrated system in real warehouse and retail store environments.This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.

对自动化库存和物品精确定位的要求已经成为现代供应链管理的关键。该项目的目标是创建一个创新的地面机器人-无人机网络系统，该系统由一组自主地面机器人和无人机组成，可在高度复杂的环境（如仓库、零售店、危险品储存设施或工厂）中提供无源射频识别（pRFID）标签物品的库存计数和精确位置。该项目将显著提高供应链管理和物联网（IoT）系统的先进水平，并为充分利用拟议的机器人-无人机平台的潜力迈出重要一步。该项目的教育计划包括发展和加强各种本科和研究生水平的课程。研究生和本科生将接触到最先进的技术，并获得在现代通信，电路和传感系统（CCSS）最前沿的尖端技术的实践经验。该项目的成果将通过技术出版物、会议报告、项目网站以及两年一度的无线工程研究与教育中心（WEREC）和RFID实验室会议进行传播。该团队完全致力于促进代表性不足的群体参与研究，并将通过NSF REU和RET计划以及与hbcu的合作等外展活动继续努力。在过去的十年中，pRFID技术被广泛应用于序列化项目级识别和数据共享。然而，大多数pRFID技术实现使用固定的读取点或人工操作的手持扫描仪，不能提供精确的物品位置。物流可视性的需求要求自动化库存和物品的精确位置信息。在本研究中，结合精确RFID定位方法的自主地面机器人-无人机网络系统将弥补上述差距。通过部署协作的地面机器人和无人机，安装商用现成的pRFID设备，提供自动库存和pRFID标签物品的精确位置。框架将异构的单个项目协作成一个连贯的系统，以完成任何单个项目都不可能完成的更复杂的任务。该架构还将为一般物联网系统提供创新的通信、控制和计算框架。该框架将作为开源工具公开，以促进CCSS社区的相关研究。以下重点将在这个项目中完成。(i)地面机器人和无人机网络架构：该架构将为地面机器人和无人机提供通信、计算和控制，以协同执行通用任务。这也为地面机器人和无人机相互配对形成共生系统提供了基本方法。（ii）地面机器人和无人机室内导航：将引入地面机器人增强机制，使无人机能够在复杂的室内环境中精确定位。当定位目标实现后，将研究一种方法，使无人机和地面机器人能够在物体丰富且受限的空间环境中安全有效地导航。（iii）准确的库存计数和pRFID的精确定位：地面机器人-无人机网络将原型化，以操作pRFID库存并提供pRFID标记物品的精确位置。（四）本项目还包括一个全面的集成和评估计划，在实际仓库和零售商店环境中测试拟议的地面机器人-无人机集成系统。该奖项反映了美国国家科学基金会的法定使命，并通过使用基金会的知识价值和更广泛的影响审查标准进行评估，被认为值得支持。

项目成果

MulTLoc: RF Hologram Tensor Filtering and Upscaling for Locating Multiple RFID Tags

• DOI: 10.1109/icccn52240.2021.9522256
• 发表时间: 2021-07
• 期刊: 2021 International Conference on Computer Communications and Networks (ICCCN)
• 作者: Xiangyu Wang;Jian Zhang;S. Mao;Senthilkumar C. G. Periaswamy;J. Patton

Acoustic-based Vital Sign Monitoring

基于声学的生命体征监测

• 发表时间: 2021
• 期刊: Proc. The 2021 IEEE-EMBS International Conference on Biomedical and Health Informatics (BHI'21
• 作者: Xuyu Wang;Shiwen Mao
• 通讯作者: Shiwen Mao

Indoor Fingerprinting With Bimodal CSI Tensors: A Deep Residual Sharing Learning Approach

• DOI: 10.1109/jiot.2020.3026608
• 发表时间: 2021-03
• 期刊: IEEE Internet of Things Journal
• 影响因子: 10.6
• 作者: Xiangyu Wang;Xuyu Wang;S. Mao

Demo Abstract: Vision-aided 3D Human Pose Estimation with RFID

演示摘要：使用 RFID 进行视觉辅助 3D 人体姿势估计

• DOI: 10.1109/msn50589.2020.00104
• 发表时间: 2020
• 期刊: Sensing and Networking (MSN 2020
• 作者: Yang, Chao;Wang, Xuyu;Mao, Shiwen
• 通讯作者: Mao, Shiwen

Temperature Forecasting for Stored Grain: A Deep Spatiotemporal Attention Approach

• DOI: 10.1109/jiot.2021.3078332
• 发表时间: 2021-12-01
• 期刊: IEEE INTERNET OF THINGS JOURNAL
• 影响因子: 10.6
• 作者: Duan, Shanshan;Yang, Weidong;Zhang, Yuan
• 通讯作者: Zhang, Yuan